I misunderstood the information.  It is only 2%.  It does not tell you that till the very end. My appologies. This is exactly what it says: UK Census Description~~~~~~~~~~~~~~~~~~~~~UKC_DESC.zip    Contains UKC_DESC.TXT --- this file which you are reading ---descriptive material for the rest of the files.UKC_NIDX.zip    227,583 bytes -- this file is an index of surnames listed inthe various county files of which the namesappear below.  The counties are given in theindex by the codes in the list of filenames.The codes used follow BS 6879:1987 which isalmost identical to the Chapman County Codes.The actual census transript files are arranged by counties.  The three lettersfollowing the underscore in the file name are the BS 6879:1987 code for thecounty to which the file relates. The following lists show the files within each compressed .zip file.  Thesizes quoted are the EXPANDED file sizes.  The naming convention for the textfiles is as follows:First three letters - BS 6879:1987 code for the countyNext two digits     - year (19th century) of the census sampledRemainder of name   - two-digit serial number --- OR ----BS 6879:1987 code for another county when a parish is inone county for registration and another geographically.All of the files are plain ASCII text files.  When there are two county codesin the main name the extension name consists of the letter "T" followed by atwo-digit serial number.  Otherwise the extension name is "TXT".The arrangement of the data within the text files is described in AlanStanier's "How to read" notes at the end of this document. The information above was prepared by Mike Fisher, FidoNet 2:440/50, from thefiles made available by Gordon Grant, FidoNet 2:253/170.Notes from Gordon Grant (transcribed by Rosemary Lockie, FidoNet 2:440/109)The transcripts were gathered together from several places by Alan Stanier.The bulk of them is a 2% sample of the 1851 census, transcribed by a teamunder Professor Anderson at Edinburgh.  Alan obtained that collection from theESRC Data Archive at Essex University.  He also obtained Eilidh Garrett's datafrom Keithley from the same Data Archive, and transcripts from an unknowncontributor of the censuses of County Antrim.  Both these had to be massagedinto the format used by Anderson.  Alan also added his own transcripts, mostlyfrom East Anglia and the Midlands.Anderson's team transcribed every 50th census book in England, Wales andScotland.  Eilidh Garrett took all the census of Keithley, as did the teamthat did County Antrim.  Alan added those he came across while tracing his ownfamily.Acknowledgments:  Professor Anderson, Eilidh Garrett, and the ESRC DataArchive, and (of course) Alan Stanier.The rest of this file consists of notes by Alan Stanier, "alan@essex.ac.uk" onthe Internet, and was extracted from Alan's messages in the Usenet newsgroup"soc.roots" and edited by Ron Macrae, FidoNet 2:440/212.Here is some information about the UK Census Data, the data fields and how toread the data.======= How to read the data below =======Census information for the United Kingdom.This file will describe the data and fields within the files listed above.OK, let's take part of a typical file.@Aberdeenshire -A Cluny#1851Z/7015/1/NETHER CORSKIE/J/AGNES/TOCHER/HEAD/W//59/FARMER OF 97 ACRES/ABD/CLUNY/J/WILLIAM/TOCHER/SON/U/28//EMPLOYED ON FARM/ABD/CLUNY/J/BETSEY/TOCHER/D/U//26/EMPLOYED ON FARM/ABD/CLUNY/J/ROBERT/TOCHER/SON/U/21//EMPLOYED ON FARM/ABD/CLUNY/J/JEAN/TOCHER/D/U//16/EMPLOYED ON FARM/ABD/CLUNY/J/GEORGE/FARQUHAR/GRAND SON/U/4///ABD/CLUNY/J/JOHN/TYTLER/SERV/U/22//FARM LAB/ABD/ECHT/J/GEORGE/BROWN/SERV/U/19//FARM LAB/ABD/SKENE/$Z/03/2/UPPER CORSKIE/J/JOHN/FINDLAY/HEAD/MAR/61//SHOE MAKER/ABD/SKENE/J/CATHERINE/FINDLAY/W/MAR//58//ABD/CLUNY/$NOTE:  If you were looking for a BROWN in the 1851 data, and the George Brownabove were to be found, then all the names of that household will be seentogether - so that you are able to see who was living with and in thehousehold of George Brown on the day of the census, and also each of theiroccupations.The first line specifies the Parish - in this case Cluny Aberdeenshire.  (The-A is not useful to you - it tells me I got the transcripts from Prof MichaelAnderson of Edinburgh University)The 2nd line (optional) gives the date of the census.Then follow a number (in this case, 2) of household records, each ended with aline starting '$'.The first line of a household record is the address line, starting with 'Z'.The first field of the address line can be ignored (it has some meaning forAnderson's research, but not for us).The second field is the household number within the enumeration book.  Thethird field (and subsequent ones if any) are the address information given inthe enumeration book.Then follow a series of person records, starting 'J'.Fields are:-Christian Names(s)SurnameRelation to head of householdMarital statusAge if male     } One of these fieldsAge if female   } should be blank!OccupationPlace of Birth (this may take more than one field, if for example botha town and a county were given.  Order is that givenin the enumeration book.  The County is coded toBS 6879:1987 - almost the same as Chapman CountyCoding)======= End of How to read =======Additional comments from Alan Stanier ...The files are my collection of transcripts, which include the 2% nationalsample.  What actually happened was I obtained the 2% sample, and was talkingto friends at the Society of Genealogists who mentioned that they wanted topublish an index to the sample, but couldn't persuade the researchers toproduce it.  So I volunteered.  I had to do a lot of data massaging first toget County names into canonical form (which, I guess, is why the researchershad declined!), so in the months I was working on that I also grabbed copiesof any other transcripts I could find.  I ended up with about 2.25% coverageof 1851, and a few from other years.The 2% sample is every 50th census book, so villages that were split betweencensus books are not fully covered.Alan Stanieralan@essex.ac.uk Renee